Sound Travels is a series of concerts and residencies for automated multi-channel diffusion of electroacoustic music. The concerts and residencies take place across Canada and around the world.

The residencies involve collaboration with a host studio or concert presenter who provides a working environment for local composers to realize multi-channel diffusion. The residencies last seven to thirty days depending on the number of composers participating and location availability.

The concerts take place immediately after the residencies in a host city, or they take place elsewhere on special invitation. The repertoire for these concerts is derived from current and/or past residencies.

Unlike the more popular forms of electronic music like techno, electroacoustic music is presented in the public as a foreground event where the context is subservient to the experience of listening to the music rather than the other way around.

There are a number of styles or branches of electroacoustic music. There are composers concerned only with realizing works in a studio, whereas there are others who devote their work to live performance. There are composers who draw on environmental sounds, there are others who generate sounds through synthesis, and there are others who use computational analysis of instrumental or environmental sounds to generate their sound palette. There are indeed further sub-categories and points of differentiation, but this is enough to provide a general impression.

Automated refers to the process of diffusion that warrants the need for the residencies. Put simply, automation involves programming computerized commands so that diffusion moves can occur hands-free. This allows the composer to concentrate completely on listening, which is important, because diffusion is a tricky business psychoacoustically speaking.

Sound Travels does not wish to turn its back on the grace and spontaneity of live musical control. Therefore it has invited David Eagle and his midi controller the aXi0 (Alternative Expressive Input Object) to join its production team. The 2001season will focus on developing diffusion strategies using both the Audiobox and aXi0 in tandem to marry the musical complexity and versatility of automated control with the finite sensitivity and responsiveness of a musical instrument.

- A sophisticated palm-rest joystick for the left hand,

- A velocity-sensitive keyboard for the right hand, and,

- Running up the performer's shoulder, an array of buttons used to change voices or trigger events (i.e. QuickTime movies, sound files, MIDI or image sequences, diffusion sequences).

In a recent work (Prelude by David Eagle) realized at the Sound Travels residency in Banff (2000), the aXi0 controlled 8-channel sound diffusion using the Audiobox and ABControl software. It quickly became apparent that this was an ideal match: by translating live performance gestures on the joystick of the aXi0 into spatial motion of sound through the surrounding loudspeakers, sound diffusion becomes part of the musical interpretation, aspects which otherwise would have been automated.

Surround sound is generally not as bleeding-edge as the marketing hype would lead one to believe. Simply wiring an extra few speakers out-of-phase with a slight delay, and perhaps adding a sub-woofer on a crossover circuit, duplicates a surround sound effect.

Soundfield microphone approaches, like Ambisonics, are more interesting, because they use a special microphone to capture the original spatial information. A soundfield microphone, an XY stereo pair, for instance, seeks to capture the original source localization, and like stereo micing, is a very effective approach. Ambisonics even takes the XY pair a step further to include the vertical or Z-axis. And in theory, Ambisonic B-format material can be mixed and edited in a relatively straightforward manner; unfortunately, commercial audio editors do not currently support the format. Additionally, although soundfield recording is effective for mimetic sources, it does not provide any short cuts or advantages in diffusing discrete, monophonic material or synthetic stereo sources.

Most localization techniques, additionally, including the traditional phantom-image created by panning between stereo pairs, are to some degree sensitive to listener position. Although there might be a sweet spot from which to appreciate the depth and breadth of the localized image, those listeners who are positioned a few degrees off axis, or a few feet closer to one speaker do not apprehend the same effect. In other cases, material is inherently impossible to localize, like low frequencies for example.

Another family of so-called 3-D sound techniques, Head-Related Transfer Functions (HRTF's) are a prime example, but are inapplicable to loudspeaker reproduction. HRTF's model the reflections of the fleshy part of the ear in order to simulate location cues; these tiny reflections are, of course completely overwhelmed when played back over loudspeakers in a conventional room.

The main point, here, is that there is no easy, one-size-fits-all solution to the problem faced by electroacoustic composers in diffusing work.

The Richmond Sound Design AudioBox, which is controlled by ABControl, supports up to 16 discrete channels (it is, after all, a 16 by 16 mixer), there's no encoding to worry about, and your material won't be savaged by lousy compression. Even more importantly, the ABControl software provides practical tools to aid in realizing diffusion (sequencing, multi-channel panning, editing, etc.). ABControl tries to fill in the gap between multi-channel audio and multi-channel composition.

Research into the psychoacoustics of localization, however, suggests many additional techniques that could be incorporated, some of which are well-studied (global reverberation vs. local reverberation, and Doppler shift, to name two).

ABControl doesn't provide any explicit support for these localization techniques. On the other hand, it doesn't prohibit you from implementing them separately, as a pre- or post-process.

Once invited, composers realizing studio-made works must arrive at the residency with an 8-channel submix of their piece that can be dubbed to a Tascam DA-88 tape machine or transferred directly to the internal harddrive of the audiobox in the format of 48khz wavfiles. When recording to DA-88 please insure that SMPTE time code is also recorded across the entire tape (see DA-88 manual for details). This 8-channel submix will then be diffused to 8 speakers using the AudioBox.

The opportunity to interact with a group of hip, savvy composers has been invaluable to Third Monk Software. There is very little applied research into multi-channel playback, but a great wealth of common practice, especially in Canada. The Sound Travels concert series is not our most lucrative venture, but it's certainly one of our most important and meaningful endeavours, and one which we will continue to support and encourage.

From Yves Daoust, compositeur, professeur-responsable du secteur électroacoustique au Conservatoire de Musique du Québec à Montréal.

Nous avons été heureux d'accueillir en janvier dernier monsieur Darren Copeland au Conservatoire de Musique de Montréal. Durant près d'un mois les étudiants de la classe d'électroacoustique du Conservatoire, auxquels se sont joints des étudiants invités des autres studios institutionnels montréalais, ont pu travailler sous la supervision de Darren Copeland à la préparation d'une diffusion multi-pistes automatisée d'une de leur oeuvre. L'atelier s'est terminé par deux concerts, l'un au Conservatoire et l'autre à l'université Concordia.

Je peux affirmer que tous les étudiants participants ont beaucoup apprécié cette Résidence" et en ont tiré un profit certain.

From Ned Bouhalassa, composer:

I have found that the residency that I undertook last November helped me to better understand what it means to compose for a multi-channel diffusion system. The freedom to test my mixes over the entire length of the residency was an added bonus. I had much experience projecting _stereo_ sound in a multi-channel system, but had not tried projecting multi-channel pieces, let alone composing directly for such a system. With independent movement of sound, it is possible to create a rich and complex sound environment where all the parts are clearly defined for the listener. The use of this system has forever changed the way that I think about composing for space in a multi-speaker context.

From Kenneth Newby, composer:

The opportunity to develop a portion of Seasonal Round in an immersive sonic environment brought the various sonic elements into a much tighter focus, heightening the drama, widening the contrasts, clarifying the spoken word. Given that the work is largely concerned with an engagement with the idea of the place of the Human in Nature, the multi-channel sound diffusion situation provided a particularly appropriate model in which to present the soundscapes that frame the spoken word.

From Lynda Hill, theatre artist:

The provision of time and equipment necessary to achieve a unique approach in spatialization within an 8-channel diffusion for the electroaacoustic portion of my sound/performance/installation piece Dark Forest meant a more effective expression of the overall intention of the work. The multi-channel spatialization of my sountrack allowed this element to become more of a kinetic compliment to the choreography of the performer and the audience's relationship to the environment of the installation.